Novel Deep Learning Model for Uncertainty Prediction in Mobile Computing

Rajawat, Anand Singh; Upadhyay, Priyanka; Upadhyay, Akhilesh R.

doi:10.1007/978-3-030-55180-3_49

Cited by 14 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This section discusses details required to implement constructive semisupervised learning (CSSL) based on the classification [ 23 ] of system spending perception of geometrical growth. To explain the algorithm for classification, the semisupervised learning [ 24 ] Algorithm is essentially constructed over multilevel geometrical. The know neurons can be extracted when it is represented by hyperspheres presentation, so they are given a learning task [ 9 ] using hyperspheres performed by the geometrical growth.…”

Section: Optimization In Support Vector Machinesmentioning

confidence: 99%

Dark Web Data Classification Using Neural Network

Rajawat

Bedi

Goyal

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

There are several issues associated with Dark Web Structural Patterns mining (including many redundant and irrelevant information), which increases the numerous types of cybercrime like illegal trade, forums, terrorist activity, and illegal online shopping. Understanding online criminal behavior is challenging because the data is available in a vast amount. To require an approach for learning the criminal behavior to check the recent request for improving the labeled data as a user profiling, Dark Web Structural Patterns mining in the case of multidimensional data sets gives uncertain results. Uncertain classification results cause a problem of not being able to predict user behavior. Since data of multidimensional nature has feature mixes, it has an adverse influence on classification. The data associated with Dark Web inundation has restricted us from giving the appropriate solution according to the need. In the research design, a Fusion NN (Neural network)-S3VM for Criminal Network activity prediction model is proposed based on the neural network; NN- S3VM can improve the prediction.

show abstract

Section: Optimization In Support Vector Machinesmentioning

confidence: 99%

Dark Web Data Classification Using Neural Network

Rajawat

Bedi

Goyal

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…(b) If the picture has a small change, the values of the particular values do not modify significantly, provided that the basic values of the image remain quite constant. [14,15] Hybrid Technique for Protection Copyright Digital image watermarking means hiding of piece of data onto an image to attest its owner or to authorize it. Digital watermark is an indicator that is hidden into a virtual media to guard it from copyright infringement.…”

Section: Dct (Discrete Cosine Transform)mentioning

confidence: 99%

Robust Digital Watermarking Techniques for Protecting Copyright

Ji¹

2021

ITII

View full text Add to dashboard Cite

Digital watermarking is the effective method of copyright defence. Typically robot-proof watermarks used to secure copyright, and are immune to some deletion or adjustment of protected documents. Fragile watermarks are typically used for content identification and are vulnerable to small alterations. In this paper, we suggest a hybrid watermarking approach that combines a solid and fragile watermark with copyright and material authentication. This mechanism is often resistant to manipulation and to clone attacks at the same time. The relationship between a delicate watermark and a stable watermark is characterised by our involvement. DCT coefficients are used to integrate the values of the watermark.

show abstract

“…Medical providers do not need a wireless tracker or an app to download to ensure that a marathon runner performs effectively to monitor the cost of calories on a day-to-day basis. [10]. Conversely, a wireless medical device used to provide a lifethreatening warning to create a degree of certainty could be needed [11].…”

mentioning

confidence: 99%

Review the Deep Learning Technique for Missing Data Classification in Iot Applications for Network Performance Improvement

Patil¹

2021

ITII

View full text Add to dashboard Cite

In order to ensure product safety and increase production quality, The construction of mine Internet of Things networks continues to accelerate mining enterprises. Given the large increase in the number of networked devices connectivity capability in the mine, there is considerable strain on the mine network communication facilities. We suggest an Innovative Solution Using Deep Learning for Missing Data Classification in IoT Network Performance Enhancement System Market Classifier based on neural networks to improve the quality of service in the connectivity infrastructure of mine networks. The classifier uses a transformation wavelet to delete the data flow and to build compliance characteristics to identify the market categories of the system.Owing to the findings of the classification, the system changes the specifications of the network services given to the terminal equipment in a versatile manner. In this way, the system's network capacity can be fairly distributed. We assess the output of the classifier model using the test data collection. We review the deep learning technique in IoT applications for Network Improvement for missing data classification

show abstract

Novel Deep Learning Model for Uncertainty Prediction in Mobile Computing

Cited by 14 publications

References 18 publications

Dark Web Data Classification Using Neural Network

Dark Web Data Classification Using Neural Network

Robust Digital Watermarking Techniques for Protecting Copyright

Review the Deep Learning Technique for Missing Data Classification in Iot Applications for Network Performance Improvement

Contact Info

Product

Resources

About